Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 May 31;65(4):1276–1292. doi: 10.1111/bjc.70069

Credibility and expectancy associated with ultra‐brief and standard‐length digital treatments: A preliminary examination of impact on treatment outcome

Melanie Seshadri 1, Blake F Dear 1, Madelyne A Bisby 1,2,✉
PMCID: PMC13655944  PMID: 42219695

Abstract

Objective

Ultra‐brief treatments (i.e., three or fewer sessions/modules) are associated with reductions in depression and anxiety symptoms, though the mechanisms underlying their effects are unclear. Research has yet to compare how perceptions of treatment credibility and outcome expectancy (i.e., how suitable and effective the treatment will be) may differ between ultra‐brief and standard‐length digital treatments.

Methods

In a secondary analysis of a randomized controlled trial (N = 152), the current study examined the distribution of credibility and expectancy prior to receiving either an ultra‐brief or standard‐length digital transdiagnostic treatment based on cognitive behaviour therapy and explored whether these variables predicted improvement in anxiety or depression symptoms.

Results

There were no differences in how credible and effective participants perceived the two treatment options. In binary logistic regressions, higher credibility and expectancy were associated with greater odds of experiencing a clinically meaningful improvement (≥ 30% reduction) in anxiety (credibility OR = 1.68 [95% CI 1.15, 2.44]; expectancy OR = 1.50 [95% CI 1.11, 2.02]) and depression symptoms (credibility OR = 1.45 [95% CI 1.02, 2.05]; expectancy OR = 1.34 [95% CI 1.03, 1.76]) for the standard‐length treatment only. The predictive value of these two variables diminished when they were entered into the models together, highlighting their conceptual overlap. Findings also indicated a potential moderating role for treatment preference.

Conclusion

Credibility and expectancy predict treatment outcomes for standard‐length digital treatment. Their influence on outcomes in ultra‐brief digital treatment is unclear and further research with larger samples is warranted.

Keywords: low mood, mental health, online, predictors, therapy

Practitioner points.

  1. Ultra‐brief treatments (e.g., a single session) can lead to reductions in anxiety and depression symptoms.

  2. At baseline, participants perceived both ultra‐brief and standard‐length digital treatment options as credible and effective.

  3. Participants who viewed the standard‐length digital treatment as more credible or effective at baseline were more likely to benefit from treatment.

INTRODUCTION

Psychological disorders rank among the leading contributors to global health burdens, and the prevalence of depression and anxiety is consistently increasing (Australian Bureau of Statistics, 2024; GBD 2021 Diseases and Injuries Collaborators, 2024). If untreated, these disorders can cause increased disability, elevated risk of suicide, reduced productivity and decreased quality of life (Cai et al., 2021; Deady et al., 2022; GBD 2021 Diseases and Injuries Collaborators, 2024). These profound impacts underscore the need for effective and accessible treatment options. Cognitive behaviour therapy is a first‐line treatment for depression and anxiety (Cuijpers et al., 2023). The digital delivery of cognitive behaviour therapy has increased its accessibility without compromising efficacy when delivered alongside therapist guidance (Hedman‐Lagerlöf et al., 2023; Kambeitz‐Ilankovic et al., 2022).

Despite the robust efficacy of both face‐to‐face and digital psychological treatments, there is a continued pattern of low use of psychological services (Bassilios et al., 2022; Hansen et al., 2002; Owen et al., 2016; Tapp et al., 2026). Cost‐related concerns are among the leading barriers to accessing psychological services, followed by time constraints, travel inconveniences, long waitlists and stigma surrounding mental health service utilization (Bartram & Stewart, 2019; Schaffler et al., 2022). Standard‐length cognitive behaviour therapy requires substantial time commitment regardless of delivery format, typically spanning at least eight sessions or online modules over several weeks to months. Addressing these barriers requires the prioritization of low‐cost and time‐limited psychological treatments that are more scalable, lower effort, and more aligned with the range of individual capacities (Cross & Alvarez‐Jimenez, 2024; Savir & Baumel, 2025).

Ultra‐brief treatments are intentionally designed to enhance accessibility by delivering psychological intervention within three or fewer sessions (Sperry & Binensztok, 2019). This treatment approach is supported not only pragmatically but also theoretically, with decades of research supporting a dose–response effect where the largest improvements occur early in treatment (Bisby et al., 2023; Howard et al., 1986; Robinson et al., 2020). A meta‐analysis of ultra‐brief treatments (k = 34; N = 3679) has indicated that these treatments result in small to moderate reductions in depression symptoms (g = 0.35) and moderate reductions in anxiety symptoms (g = 0.51) relative to control conditions (e.g., waitlist, treatment as usual) at post‐treatment (Bisby et al., under review). Similar findings were reported in an umbrella review of single‐session interventions (k = 24; N = 40,629), in which these interventions were associated with reductions in youth anxiety (g = 0.52), depression (g = 0.12) and substance use outcomes (g = 0.25) (Schleider et al., 2025). For some adults with anxiety or depression, ultra‐brief digital treatments may be no less effective than standard‐length digital treatments after 2 months (Bisby et al., 2024); however, further research directly comparing these treatment modalities is critical. Nevertheless, ultra‐brief treatments represent a much‐needed pathway to increasing the accessibility of psychological treatments.

It remains unclear who benefits most from ultra‐brief treatments, and how these treatments exert their effects. Developing a mechanistic understanding of ultra‐brief treatments would ultimately help refine existing protocols and facilitate the development of more targeted, effective and accessible psychological treatments. The contextual model provides a framework for understanding the mechanisms of psychological treatments by acknowledging the interaction between specific factors (i.e., information, skills and techniques unique to a therapeutic approach) and common treatment factors (i.e., shared elements across all therapeutic approaches) (Cuijpers et al., 2019; Wampold & Imel, 2015). Credibility (perceptions of treatment suitability and plausibility) and expectancy (the extent to which an individual believes that the treatment will result in symptom improvement) are two common factors which have been identified as contributing to treatment outcomes. In a comprehensive secondary analysis of randomized controlled trials of digital standard‐length treatment for depression and anxiety disorders, Dear (2025) examined the distributions of credibility and expectancy at baseline for the first time. Most participants reported relatively high credibility and expectancy in this sample, indicating a possible self‐selection bias where individuals with favourable expectations around treatment are more likely to sign up for digital treatments. This study also demonstrated that higher average credibility and expectancy scores were associated with lower dropout, higher treatment completion and larger reductions in depression and anxiety symptoms, highlighting the importance of understanding the impact of these constructs on treatment outcomes.

The influence of credibility and expectancy in ultra‐brief treatments has only been reported in one study to date. In a secondary analysis of a randomized controlled trial of a three‐lesson digital treatment for managing rumination and worry (N = 137), Upton et al. (2025) explored baseline predictors of reductions in depression and anxiety symptoms. Higher baseline credibility and expectancy (entered as a combined score) predicted lower depression and anxiety scores at post‐treatment, even after considering the impacts of baseline symptom severity and treatment group (clinician vs. self‐guided). This study suggests that credibility and expectancy are also associated with symptom change in ultra‐brief digital treatments.

Although conceptually related and highly correlated, credibility and expectancy are considered distinct factors (Dear, 2025; Luong et al., 2020; Thompson‐Hollands et al., 2014). Meta‐analytic evidence indicates that higher credibility and expectancy are generally associated with better treatment outcomes, although the sample size of meta‐analyses to date has been relatively modest (Kumpasoğlu et al., 2025; Pontén et al., 2024). An individual participant data meta‐analysis (k = 6, N = 491) of randomized controlled trials comparing face‐to‐face and digital standard‐length treatments reported that there was no difference in expectations across treatment modalities, and higher expectancy significantly predicted symptom improvements (β = 0.27) irrespective of modality (Pontén et al., 2024). These findings suggest that the impact of expectations on outcomes is consistent across how treatments are delivered. Nevertheless, it remains unclear whether treatments of different intensities or durations are perceived as equally credible and effective.

The current study aimed to address critical gaps in the literature by examining credibility and expectancy relating to a digital ultra‐brief treatment for depression and anxiety. By utilizing existing data from a non‐inferiority randomized controlled trial (Bisby et al., 2024), we were able to include a digital standard‐length treatment for depression and anxiety as a comparator. The study had two aims: (1) to examine the distributions of credibility and expectancy prior to starting digital ultra‐brief or standard‐length treatment, and (2) to examine whether higher baseline credibility or expectancy predicted clinically meaningful symptom improvements. It was hypothesized that the current study would replicate existing literature, with relatively high baseline credibility and expectancy scores. It was also hypothesized that higher baseline credibility and expectancy would predict greater symptom improvement in both the standard‐length and ultra‐brief treatment groups. Following the initial univariate models, hierarchical binary logistic regression analyses were used to examine the effects of credibility and expectancy after considering other clinically relevant predictors of outcome. These predictors included baseline anxiety and depression symptoms (Scholten et al., 2023; Sextl‐Plötz et al., 2024), whether they received their preferred treatment option (Bennett et al., 2025), recent mental health treatment (Reins et al., 2019) and concurrent medication for mental health (Staples et al., 2022).

METHODS

Design

The current study is a secondary analysis of a three‐arm randomized controlled trial (N = 242) comparing a digital ultra‐brief treatment, a digital standard‐length treatment and a waitlist control in adults with depression or anxiety (Bisby et al., 2024). Only treatment group participants with complete baseline data were included, resulting in a sample of 152 participants (ultra‐brief n = 77, standard‐length n = 75). Baseline refers to pre‐treatment, which occurred in the first week of treatment (noting that participants could not access the treatment until after the pre‐treatment questionnaires had been completed). Before completing the credibility and expectancy ratings, participants had viewed a webpage and a consent form outlining the treatment's purpose, content, structure, suitability and supporting evidence.

Participants

Participants were recruited through social media advertising and the eCentreClinic website (ecentreclinic.org). Interested and consenting individuals completed an online assessment application followed by a telephone assessment with a psychologist to confirm eligibility. Participants were included if they were (1) an Australian resident, (2) aged 18 years or over and (3) experiencing difficulties with anxiety or depression, as indicated by participant self‐report. Participants were considered ineligible if they were (1) living outside of Australia, (2) unable to read or understand English, (3) actively suicidal or (4) unable to keep themselves safe.

Treatment

All participants were supported by a clinical psychologist. Therapist contact was optional and available via phone call or private messaging. Participants in the ultra‐brief treatment had one week to contact the psychologist after completing the lesson, while those in the standard‐length treatment could contact the psychologist for the duration of their eight‐week course.

Ultra‐brief treatment

The treatment included one online lesson (45 slides), practice exercises, illustrative case stories and an additional resource on problem‐solving. The content was based on cognitive behaviour therapy; the lesson was very similar to the first lesson of the standard‐length treatment. Further details can be found in the primary publication (Bisby et al., 2024).

Standard‐length treatment

The treatment included five online lessons (31–43 slides), practice exercises, illustrative case stories and seven additional resources. The treatment is based on cognitive behaviour therapy. Further details can be found in previous publications (Dear et al., 2011; Titov et al., 2015).

Outcomes

Generalized Anxiety Disorder—7‐item (GAD‐7)

The GAD‐7 is a 7‐item self‐report measure of anxiety symptoms in the previous 2 weeks (Spitzer et al., 2006). Items range from 0 (not at all) to 3 (nearly every day), the total score ranges from 0 to 21, and the measure demonstrates acceptable psychometric properties (Johnson et al., 2019; Plummer et al., 2016). Cronbach's alpha = .85.

Patient Health Questionnaire—9‐item (PHQ‐9)

The PHQ‐9 is a 9‐item self‐report measure of depression symptoms in the previous 2 weeks (Kroenke et al., 2001). Items range from 0 (not at all) to 3 (nearly every day) and the total score ranges from 0 to 27. The PHQ9 shows acceptable psychometric properties (Levis et al., 2019; Titov et al., 2011). Cronbach's alpha = .83.

Credibility and Expectancy Questionnaire (CEQ)

The CEQ is a 6‐item measure of treatment credibility and expectancy (Devilly & Borkovec, 2000). The first three items relate to credibility (e.g., ‘how logical does the treatment seem?’) and are scored from 1 to 9, resulting in a total score range of 3 to 27. The second three items relate to expectancy (e.g., ‘How much improvement do you think will occur?’). Items 4 and 6 are scored from 0% to 100% in 10% increments, whereas item 5 is scored from 1 to 9. To ensure consistency and comparability, expectancy items 4 and 6 were rescaled to a 0–10 metric, resulting in a total score of 1 to 29. For interpretability, average credibility and expectancy item scores are reported. Cronbach's alpha ranged from 0.78 (credibility) to 0.89 (expectancy).

Statistical analysis

All analyses were performed using IBM SPSS version 29 with an alpha of .05. Descriptive statistics were used to examine the distribution of credibility and expectancy scores between treatment groups. Skewness and kurtosis in the range of −2 to +2 were considered within an acceptable range. Multicollinearity was not a concern; all VIF values were below five. The Kolmogorov–Smirnov test was used to test whether credibility and expectancy scores for each group were normally distributed. In both cases, the assumption of normality was violated (largest p = .045) and therefore the Mann–Whitney U test was used to compare scores between groups.

Prior to regression analyses, multiple imputation was used to handle missing post‐treatment data (15% missing) using group, lesson completion, credibility, expectancy, baseline symptom severity, mental health treatment history and medication as predictors of missingness (Karin et al., 2021). Separate binary logistic regression analyses were conducted for each treatment group. Binary logistic regression analyses using the complete case sample are presented in the Supporting Information. Univariate models first examined whether higher baseline credibility or expectancy predicted significantly higher odds of experiencing a clinically meaningful improvement in depression or anxiety symptoms at post‐treatment (Bisby et al., 2023). Clinically meaningful improvement was operationalized as a ≥ 30% reduction in symptoms from baseline, consistent with previous literature indicating that this level is noticeably impactful for individuals and is recommended over other dichotomous methods of responder analysis (Smith et al., 2020; van Zwet et al., 2026). Subsequent hierarchical binary logistic regression analyses were conducted to examine the effect of credibility and expectancy when accounting for other identified predictors of treatment outcome. We report model fit using the −2LL statistic, in which lower values indicate better model fit. The null model included baseline depression symptoms, baseline anxiety symptoms, treatment preference (received preferred, received not preferred, neutral), mental health treatment in the past 12 months (yes/no) and current medication for mental health (yes/no). Credibility was entered in the second block and expectancy in the third block. A reverse model was also examined (i.e., block 2: expectancy, block 3: credibility). Changes in model fit between blocks are reported. Odds ratios with 95% confidence intervals are reported.

RESULTS

Sample characteristics

Participant age ranged from 22 to 78 years old (M = 47.90, SD = 12.83). Most participants were female (86%) and lived in a capital city or its surrounds (72%). Most participants had at least an undergraduate degree (72%), were in full‐time paid employment (43%) and had received mental health treatment in the past 12 months (53%). Less than half of the participants were taking medication for their mental health (42%). Treatment preference was evenly distributed across groups, with 29% preferring the ultra‐brief treatment, 36% preferring the standard‐length treatment and 36% reporting no preference. There were no significant group differences in demographic or clinical characteristics (ps > .05; see Table 1).

TABLE 1.

Sample characteristics.

Total sample N = 152 Ultra‐brief N = 77 Standard‐length N = 75
Age (M, SD) 47.90 (12.83) 49.65 (11.71) 46.11 (13.74)
Baseline GAD‐7 (M, SD) 9.72 (4.51) 10.26 (4.75) 9.16 (4.20)
Baseline PHQ‐9 (M, SD) 11.21 (5.14) 11.69 (5.13) 10.73 (5.14)
Gender
Female 130 (86) 69 (90) 61 (81)
Male 21 (14) 7 (9) 14 (19)
Other 1 (1) 1 (1) 0 (0)
Location
Capital city or surrounds 110 (72) 52 (68) 58 (77)
Other urban region 23 (15) 15 (19) 8 (11)
Rural or remote 19 (13) 10 (13) 9 (12)
Education
Year 12 or less 14 (9) 3 (4) 11 (15)
Trade certificate or apprenticeship 10 (7) 5 (6) 5 (7)
Undergraduate or associate diploma 18 (12) 10 (13) 8 (11)
Undergraduate degree 78 (51) 46 (60) 32 (43)
Post‐graduate degree 32 (21) 13 (17) 19 (25)
Employment
Full‐time paid 72 (43) 33 (38) 39 (48)
Part‐time paid 37 (22) 21 (24) 16 (20)
Casual 19 (11) 12 (14) 7 (9)
Student 0 (0) 0 (0) 0 (0)
At home parent 10 (6) 6 (7) 4 (5)
Unemployed or seeking work 9 (5) 3 (3) 6 (7)
Registered sick or disabled 11 (7) 6 (7) 5 (6)
Retired 5 (3) 3 (3) 2 (3)
At home not seeking work 6 (4) 4 (5) 2 (3)
Preference
Ultra‐brief 44 (29) 23 (30) 21 (28)
Standard‐length 54 (36) 27 (35) 27 (36)
Either 54 (36) 27 (35) 27 (36)
Recent mental health treatment
No 72 (47) 37 (48) 35 (47)
Yes 80 (53) 40 (52) 40 (53)
Medication for mental health
No 88 (58) 42 (54) 46 (61)
Yes 64 (42) 35 (46) 29 (39)
Open in a new tab

Note: N (%) reported. Participants could endorse more than one employment option.

Distribution of baseline credibility and expectancy

The distribution of baseline credibility and expectancy scores is presented in Figures 1 and 2. Skewness and kurtosis were within guidelines (see Table 2). Assumption testing indicated non‐normality using the Kolmogorov–Smirnov test, and therefore, the Mann–Whitney U test was used to compare distributions across groups. Results indicated no significant differences in credibility (U = 3019, Z = .486, p = .63) or expectancy (U = 2594, Z = −1.084, p = .27) between those receiving the ultra‐brief or standard‐length treatments.

FIGURE 1.

FIGURE 1

Open in a new tab

Distribution of credibility prior to (a) ultra‐brief and (b) standard‐length digital treatment. Rounded average scores reported.

FIGURE 2.

FIGURE 2

Open in a new tab

Distribution of average expectancy prior to (a) ultra‐brief and (b) standard‐length digital treatment. Rounded average scores reported.

TABLE 2.

Descriptive statistics.

Credibility Expectancy
Ultra‐brief Standard‐length Ultra‐brief Standard‐length
Mean 6.03 5.93 4.23 4.55
Standard deviation 1.41 1.53 1.93 1.92
Median 6.00 6.00 4.00 4.67
IQR 1.83 2.00 2.67 2.33
25th percentile 5.00 5.00 3.00 3.33
75th percentile 6.67 7.00 5.67 5.67
Skewness −0.08 0.03 0.38 0.05
Kurtosis 0.21 −0.21 −0.58 −0.47
Open in a new tab

Abbreviation: IQR, Interquartile range.

Effects of baseline credibility and expectancy on the occurrence of clinical change

Ultra‐brief treatment

Across the sample, 58% [95% CI: 57, 59] and 58% [95% CI: 57, 58] reported a clinically meaningful improvement in anxiety and depression, respectively. Average credibility was not significantly associated with the odds of experiencing a clinically meaningful improvement in anxiety (OR = 1.00 [95% CI 0.69, 1.45], p = .98) or depression (OR = 1.19 [95% CI 0.82, 1.72], p = .36). Similarly, average expectancy was not significantly associated with the odds of clinically meaningful improvement in anxiety (OR = 1.04 [95% CI 0.81, 1.33], p = .78) or depression (OR = 1.24 [95% CI 0.95, 1.63], p = .12) for participants in the ultra‐brief treatment group. Results using the complete case sample were consistent (see Supporting Information).

Standard‐length treatment

Across the sample, 50% [95% CI: 50, 50] and 51% [95% CI: 51, 51] reported a clinically meaningful improvement in anxiety and depression, respectively. In contrast, higher average credibility was associated with increased odds of experiencing a clinically meaningful improvement in anxiety (OR = 1.68 [95% CI 1.15, 2.44], p = .007) and depression (OR = 1.45 [95% CI 1.02, 2.05], p = .04). In other words, each one‐unit increase in average credibility was associated with 68% higher odds of improvement in anxiety and 45% higher odds of improvement in depression. Higher average expectancy was associated with increased odds of experiencing a clinically meaningful improvement in anxiety (OR = 1.50 [95% CI 1.11, 2.02], p = .008) and depression (OR = 1.34 [95% CI 1.03, 1.76], p = .03). These results suggest that a one‐unit increase in average expectancy was associated with 50% higher odds of improvement in anxiety and 34% higher odds of improvement in depression. All complete case analyses were consistent with the primary analyses using imputed data (see Supporting Information).

Effects of baseline credibility and expectancy on the occurrence of clinical change: Hierarchical models

Ultra‐brief treatment: anxiety symptoms

In the null model (−2LL = 94.66), higher baseline anxiety symptoms were associated with an increased likelihood of experiencing a clinically meaningful improvement in anxiety symptoms (OR = 1.19 [95% CI 1.00, 1.42], p = .049; see Table 3). Baseline depression severity, preference, mental health treatment and medication use were non‐significant. The additions of credibility (block 2) and expectancy (block 3) did not significantly improve the model (block 2 −2LL 94.16, block 3 −2LL 93.13). In the final model, higher baseline anxiety remained the only significant predictor. This pattern remained the same when the addition of credibility (block 3) and expectancy (block 2) was reversed. The same pattern of findings was reported in the complete case analysis.

TABLE 3.

Multivariate binary logistic regression output for anxiety symptom models.

Ultra‐brief Standard‐length
Block 1 Block 2 Block 3 Block 1 Block 2 Block 3
Baseline credibility –

1.09

(0.69, 1.72)

0.93

(0.50, 1.73)

 –

1.60

(1.05, 2.44)*

1.38

(0.82, 2.32)
Baseline expectancy – –

1.20

(0.80, 1.80)

 – –

1.29

(0.87, 1.92)
Baseline anxiety symptoms

1.19

(1.00, 1.42)*

1.21

(1.00, 1.46)

1.22

(1.01, 1.48)*

1.04

(0.91, 1.20)

1.05

(0.91, 1.21)

1.04

(0.90, 1.20)
Baseline depression symptoms

0.90

(0.76, 1.06)

0.89

(0.75, 1.06)

0.89

(0.75, 1.06)

0.93

(0.83, 1.03)

0.91

(0.81, 1.03)

0.91

(0.81, 1.03)
Preference
Preferred ref ref ref ref ref ref
Not preferred

0.52

(0.14, 1.99)

0.52

(0.14, 2.00)

0.50

(0.13, 1.94)

0.27

(0.07, 0.98)*

0.46

(0.11, 1.90)

0.51

(0.12, 2.12)
Neutral

1.33

(0.36, 4.83)

1.29

(0.35, 4.81)

1.27

(0.34, 4.77)

0.71

(0.23, 2.22)

0.86

(0.26, 2.86)

0.83

(0.26, 2.96)
Previous mental health treatment
No ref ref ref ref ref ref
Yes

1.36

(0.48, 3.82)

1.41

(0.49, 4.01)

1.40

(0.49, 4.02)

1.15

(0.42, 3.16)

1.30

(0.45, 3.72)

1.44

(0.49, 4.23)
Medication
No ref ref ref ref ref ref
Yes

2.07

(0.62, 6.91)

2.04

(0.60, 6.93)

2.15

(0.63, 7.28)

0.88

(0.30, 2.52)

0.90

(0.30, 2.72)

0.80

(0.28, 2.67)
Open in a new tab

Note: Odds ratios with 95% confidence intervals reported. *p < .05.

Abbreviation: ref., reference group.

Standard‐length treatment: anxiety symptoms

In the null model (−2LL = 95.97), not receiving the preferred treatment was the only significant predictor of a lower likelihood of improvement in anxiety (OR = 0.27 [95% CI 0.07, 0.98], p = .046; see Table 3). Baseline anxiety symptoms, baseline depression symptoms, preference, recent mental health treatment and medication use were not significant. The addition of credibility into the model significantly improved model fit (−2LL = 89.42; p = .01) and higher credibility predicted higher odds of improvement (OR = 1.60 [95% CI 1.05, 2.44], p = .03). The addition of expectancy did not improve model fit (−2LL = 87.05), the effect of credibility was diminished, and no predictors were significant in the final model. When the order was reversed, the addition of expectancy in block 2 significantly improved model fit (p = .01) and higher expectancy was associated with greater odds of improvement in anxiety (OR = 1.45 [95% CI 1.06, 1.99], p = .02). The addition of credibility did not improve model fit, and no predictors were significant in the final model when reversed. The same pattern of findings was reported in the complete case analysis.

Ultra‐brief treatment: depression symptoms

In the null model (−2LL = 90.97), having a neutral preference was associated with higher odds of experiencing a clinically meaningful improvement in depression symptoms (OR = 6.07 [95% CI 1.44, 25.70], p = .01; see Table 4). Baseline anxiety symptoms, baseline depression symptoms, recent mental health treatment and medication use were not significant. The additions of credibility (block 2) and expectancy (block 3) did not significantly improve the model (block 2 −2LL = 90.61; block 3 −2LL = 89.35). In the final model, having a neutral treatment preference remained the only significant predictor of greater odds. This pattern was the same in the reverse model and when using the complete case sample.

TABLE 4.

Multivariate binary logistic regression output for depression symptom models.

Ultra‐brief Standard‐length
Block 1 Block 2 Block 3 Block 1 Block 2 Block 3
Baseline credibility –

1.08

(0.70, 1.67)

0.91

(0.50, 1.64)

 –

1.54

(1.01, 2.35)*

1.26

(0.74, 2.13)
Baseline expectancy – –

1.23

(0.81, 1.87)

 – –

1.36

(0.91, 2.04)
Baseline anxiety symptoms

0.88

(0.74, 1.07)

0.90

(0.74, 1.09)

0.91

(0.75, 1.11)

0.89

(0.77, 1.05)

0.90

(0.77, 1.06)

0.88

(0.75, 1.03)
Baseline depression symptoms

1.06

(0.89, 1.27)

1.06

(0.88, 1.27)

1.05

(0.88, 1.27)

1.14

(1.01, 1.30)*

1.15

(1.01, 1.31)*

1.18

(1.02, 1.35)*
Preference
Preferred ref ref ref ref ref ref
Not preferred

1.07

(0.23, 4.95)

1.07

(0.23, 4.91)

1.04

(0.22, 5.00)

0.74

(0.18, 2.95)

1.40

(0.28, 6.98)

1.61

(0.32, 7.98)
Neutral

6.07

(1.44, 25.70)*

5.93

(1.41, 24.95)*

5.75

(1.34, 24.66)*

2.44

(0.71, 8.42)

3.24

(0.88, 11.89)

3.38

(0.89, 12.92)
Previous mental health treatment
No ref ref ref ref ref ref
Yes

1.09

(0.27, 4.40)

1.12

(0.28, 4.50)

1.10

(0.27, 4.54)

1.28

(0.44, 3.77)

1.44

(0.47, 4.40)

1.62

(0.52, 5.10)
Medication
No ref ref ref ref ref ref
Yes

0.64

(0.14, 2.91)

0.63

(0.14, 2.92)

0.67

(0.14, 3.27)

0.40

(0.12, 1.35)

0.38

(0.10, 1.39)

0.33

(0.09, 1.23)
Open in a new tab

Note: Odds ratios with 95% confidence intervals reported. *p < .05.

Abbreviation: ref., reference group.

In the null model (−2LL = 93.20), higher baseline depression symptoms were significantly associated with higher odds of experiencing a clinically meaningful improvement in depression (OR = 1.14 [95% CI 1.01, 1.30], p = .03; see Table 4). Baseline anxiety, preference, recent mental health treatment and medication use were not significant. The addition of credibility into the model significantly improved model fit (−2LL = 87.78; p = .01) and higher credibility predicted higher odds of experiencing an improvement in depression (OR = 1.54 [95% CI 1.01, 2.35], p = .04). When expectancy was added, the effect of credibility was diminished (−2LL = 84.86). Higher baseline depression symptoms remained the only significant predictor in the final model. When the order was reversed, the addition of expectancy in block 2 significantly improved model fit (p = .007) and higher expectancy was associated with greater odds of experiencing a clinically meaningful improvement in depression (OR = 1.50 [95% CI 1.09, 2.07], p = .01). Higher baseline depression symptoms continued to predict greater odds of an improvement in depression symptoms in the final model. This pattern did not change in the reverse model. In the complete case analysis, only baseline depression symptoms predicted clinically meaningful improvement in depression; this occurred in all three blocks.

DISCUSSION

The present study examined the role of baseline credibility and expectancy in predicting improvements in anxiety and depression symptoms following ultra‐brief or standard‐length digital treatments for anxiety and depression. Average credibility and expectancy scores did not differ significantly between treatment groups and followed a mostly normal distribution, noting that one statistical test did detect non‐normality for both variables. Unexpectedly, in univariate models, higher baseline credibility and expectancy predicted higher odds of experiencing a clinically meaningful improvement in depression and anxiety symptoms for the standard‐length group only. Credibility and expectancy continued to predict improvements for the standard‐length group when other known predictors of outcome were also considered (baseline symptom severity, treatment preference, treatment history and medication use), but not when they had both been entered into the same models, highlighting the conceptual overlap between these variables. These findings emphasize the need for further research with larger samples to explore the influence of credibility and expectancy on treatment outcomes and how they may differ between treatment formats.

Average credibility and expectancy ratings did not differ between treatment groups, with most participants anticipating a 40%–50% improvement across treatment formats. It can be challenging to compare our results to previous literature due to differences in the transformation and scaling of CEQ scores. (e.g., reporting total scores: Bernstein et al., 2023; or reporting item scores: Joubert et al., 2023). The present findings are consistent with those of Dear (2025), who employed similar transformation and reporting (average credibility M = 6.33; average expectancy M = 5.11). Furthermore, most participants rated the treatment as highly credible (i.e., average scores above 5), suggesting potential under‐sampling of the general population. Specifically, those individuals who perceived the treatment as less credible may have been less likely to sign up for the treatment and are therefore underrepresented in the distribution. The source randomized controlled trial for the current study (Bisby et al., 2024) advertised, described and promoted the ultra‐brief digital treatment as potentially no less effective than the standard‐length digital treatment in reducing anxiety and depression symptoms. It is likely that this approach and framing resulted in largely equivalent perceptions of how credible and effective the two treatment options were and this may have also contributed to the non‐inferiority results observed in the original study. Future studies which manipulate how treatment options are framed may shed light on this important consideration.

As expected, higher baseline credibility and expectancy both predicted increased odds of experiencing improvements in anxiety and depression symptoms for the standard‐length treatment group. This finding is consistent with past research, which has reported positive associations between credibility, expectancy and outcomes of standard‐length treatments (Dear, 2025; Kumpasoğlu et al., 2025; Luong et al., 2020; Pontén et al., 2024). However, neither baseline credibility nor expectancy predicted the occurrence of clinically meaningful improvements in anxiety or depression symptoms in the ultra‐brief treatment group. This finding contrasts Upton et al. (2025), who reported that higher baseline total CEQ scores were associated with lower depression and anxiety at post‐treatment following an ultra‐brief digital treatment for adults managing rumination and worry. There are several possible explanations for these discrepant findings: (1) the current study utilized a shorter treatment protocol [1‐lesson vs. 3‐lesson], (2) the current study examined credibility and expectancy as separate CEQ subscales rather than a combined total CEQ score or (3) the current sample may not have been large enough to detect the effects observed by Upton et al. (2025) using a larger sample (N = 137 vs. N = 77). The last suggestion is supported by the observation that the confidence intervals for the effects of credibility and expectancy on treatment outcomes for the standard‐length and ultra‐brief treatment groups overlapped in the current study. Clearly, future studies with larger samples are needed to better explore the predictive value of these constructs under different treatment conditions.

In the standard‐length treatment group, we observed that credibility and expectancy predicted improvements when only one of the two variables was included in the hierarchical binary logistic regressions, but this effect was diminished when the other was also included. A similar finding was reported by Smeets et al. (2008) using a sample of adults with chronic pain (n = 167) who received rehabilitation treatment. In that study, outcome expectancy predicted improvements in self‐reported disability only when credibility had not been controlled for in the model. The authors attributed this finding to the overlap in the two constructs: although they are distinct (as indicated by the lack of multicollinearity), they share significant variance. Consequently, entering both variables into the hierarchical models may have obscured the predictive value of these variables without implying clinical irrelevance. Future research should utilize structural equation modelling and larger samples to better understand how these factors interact or overlap in predicting treatment outcomes.

Higher baseline anxiety symptoms emerged as the only predictor of experiencing a clinically meaningful improvement in anxiety symptoms in the ultra‐brief treatment group. This is broadly consistent with literature illustrating that higher baseline anxiety is associated with greater reductions in anxiety (e.g., Bisby et al., 2022; Chandra et al., 2025; Scholten et al., 2023). At the same time, higher baseline depression symptoms consistently predicted greater odds of experiencing a clinically meaningful improvement in depression symptoms for the standard‐length treatment group. Importantly, as noted above, several of the confidence intervals of the associations between these predictors and symptom change overlapped between ultra‐brief and standard‐length treatment groups, indicating that it is unclear whether the association between baseline symptoms and outcome is moderated by treatment length or intensity.

Having a neutral preference emerged as a significant predictor of improvements in depression symptoms for the ultra‐brief treatment group. Specifically, those participants who reported no preference between the treatment formats were six times more likely to report an improvement in depression compared to those who received the treatment they preferred (i.e., matched). This finding contrasts with what was observed in the standard‐length group, in which not receiving the preferred treatment was associated with lower odds of improvement in anxiety symptoms. Indeed, this latter finding was more consistent with the existing literature, which suggests that receiving the preferred treatment (versus receiving the not preferred treatment) is associated with small but significant improvements in outcomes (k = 46: Bennett et al., 2025). To date, the field has largely operationalized preference as dichotomous, either forcing an explicit choice between treatment options or assigning participants to their preferred or not preferred treatment (e.g., Cooper et al., 2018; Linardon et al., 2023). This dichotomy has overlooked the impacts of having a neutral preference, and how this may impact treatment engagement and outcomes. It is also worth noting the general pattern of our findings—that receiving the not preferred treatment was disadvantageous for anxiety outcomes and that having a neutral preference was advantageous for depression outcomes. Future research is encouraged to replicate our findings and, if replicated, examine the mechanisms by which distinct preferences confer benefits or risks for different treatment outcomes.

The findings of this study should be interpreted alongside several limitations. First, all participants in the original randomized controlled trial were willing to be randomized to receive an ultra‐brief digital treatment for managing anxiety and depression symptoms; this may limit the generalisability of the current findings to broader populations with different motivations and treatment preferences. Second, we standardized items four and six of the CEQ due to the different scale ranges and used the average score (Dear, 2025). This reflects a broader limitation of the field in which studies handle CEQ data in different ways and report different total, average or item scores (e.g., Berg et al., 2025; Joubert et al., 2023; Newman & Fisher, 2010; Silva et al., 2021). Future studies could develop standardized transformation guidelines to enable cross‐study comparisons. Third, the sample size of the original study may have prevented the current analysis from detecting smaller predictive effects or treatment group differences, highlighting the importance of future studies having larger sample sizes. Fourth, the current study only examined one clinical outcome, defined as a ≥ 30% improvement in anxiety or depression symptoms from pre‐treatment to post‐treatment. Future research could examine a range of outcomes, including the general moderation of change in symptoms over time, as well as treatment drop‐out and engagement (Dear, 2025).

Our findings contribute to the developing literature concerning moderators of response to digital treatments and how they may differ based on intensity. First, our results indicate that individuals who are interested in ultra‐brief digital treatments may perceive them to be as credible and effective as a comparative standard‐length digital treatment when they are advertised in that manner. These constructs only predicted treatment outcome for the standard‐length group (noting that we may have been insufficiently powered to detect potentially smaller effects on outcome in the ultra‐brief treatment). The moderating influence of credibility and expectancy on the engagement and efficacy of standard‐length treatments has been consistently demonstrated (Dear, 2025; Kumpasoğlu et al., 2025; Pontén et al., 2024), suggesting that designing and delivering credibility or expectation‐enhancing interventions prior to treatment could bolster treatment engagement and outcome, at least for standard‐length treatments. This study was designed as a first step towards understanding mechanisms of symptom change in ultra‐brief versus standard‐length digital treatments; further research exploring the influence of credibility and expectancy on engagement and outcome in ultra‐brief treatments is warranted.

AUTHOR CONTRIBUTIONS

Melanie Seshadri: Conceptualization; investigation; writing – original draft; methodology; writing – review and editing. Blake F. Dear: Data curation; supervision; resources; writing – review and editing; methodology; validation. Madelyne A. Bisby: Conceptualization; investigation; writing – review and editing; funding acquisition; visualization; formal analysis; project administration; data curation; supervision; resources; validation.

FUNDING INFORMATION

This research was funded by a Macquarie University Research Fellowship to MB.

CONFLICTS OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Table S1. Univariate binary logistic regression output with complete case data.

Table S2. Multivariate binary logistic regression output for anxiety symptom models with complete case data.

Table S3. Multivariate binary logistic regression output for depression symptom models with complete case data.

BJC-65-1276-s001.docx (23.1KB, docx)

ACKNOWLEDGEMENTS

Open access publishing facilitated by Macquarie University, as part of the Wiley ‐ Macquarie University agreement via the Council of Australasian University Librarians.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author pending ethical approval and institutional agreement. The data are not publicly available due to privacy or ethical restrictions.

REFERENCES

  1. Australian Bureau of Statistics . (2024). National Study of Mental Health and Wellbeing, 2020–2022. https://www.abs.gov.au/statistics/health/mental‐health/national‐study‐mental‐health‐and‐wellbeing/latest‐release
  2. Bartram, M. , & Stewart, J. M. (2019). Income‐based inequities in access to psychotherapy and other mental health services in Canada and Australia. Health Policy (Amsterdam, Netherlands), 123(1), 45–50. 10.1016/j.healthpol.2018.10.011 [DOI] [PubMed] [Google Scholar]
  3. Bassilios, B. , Ftanou, M. , Machlin, A. , Mangelsdorf, S. , Tan, A. , Scurrah, K. , Morgan, A. , Roberts, L. , Banfield, M. , Spittal, M. , Mihalopoulos, C. , & Pirkis, J. (2022). Independent evaluation of supported digital mental health services – Phase 2 final report.
  4. Bennett, C. B. , Toevs, E. K. , Angus, A. , Krawczak, R. , Trentacosta, C. J. , Jagadeesh, D. , Tomlinson, C. S. , & Lindhiem, O. (2025). Meta‐analysis on the effects of client preferences on treatment satisfaction, completion, and clinical outcome: A 10‐year update. Clinical Psychology Review, 121, 102644. 10.1016/j.cpr.2025.102644 [DOI] [PubMed] [Google Scholar]
  5. Berg, H. , Tjelle, K. , Hansen, B. , Solem, S. , Björgvinsson, T. , Kvale, G. , & Hagen, K. (2025). Treatment expectancy and credibility as predictors of concentrated exposure treatment outcomes in patients with difficult‐to‐treat obsessive‐compulsive disorder. BMC Psychiatry, 25, 275. 10.1186/s12888-025-06737-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bernstein, E. E. , Weingarden, H. , Greenberg, J. L. , Williams, J. , Hoeppner, S. S. , Snorrason, I. , Phillips, K. A. , Harrison, O. , & Wilhelm, S. (2023). Credibility and expectancy of smartphone‐based cognitive behavioral therapy among adults with body dysmorphic disorder. Journal of Obsessive‐Compulsive and Related Disorders, 36, 100781. 10.1016/j.jocrd.2023.100781 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bisby, M. A. , Balakumar, T. , Scott, A. J. , Titov, N. , & Dear, B. F. (2024). An online therapist‐guided ultra‐brief treatment for depression and anxiety: A randomized controlled trial. Psychological Medicine, 54(5), 902–913. 10.1017/S003329172300260X [DOI] [PubMed] [Google Scholar]
  8. Bisby, M. A. , Fisher, A. , Jervis, N. , Gandy, M. , Chandra, S. S. , Heriseanu, A. I. , Hathway, Taylor, Correa, A. B. , Titov, N. , & Dear, B. F. (under review). Ultra‐brief psychological treatment for adults with depression or anxiety symptoms: A systematic review and meta‐analysis. Cognitive Behaviour Therapy. [DOI] [PubMed] [Google Scholar]
  9. Bisby, M. A. , Scott, A. J. , Fisher, A. , Gandy, M. , Hathway, T. , Heriseanu, A. I. , Karin, E. , Cross, S. , Staples, L. , Titov, N. , & Dear, B. F. (2023). The timing and magnitude of symptom improvements during an internet‐delivered transdiagnostic treatment program for anxiety and depression. Journal of Consulting and Clinical Psychology, 91(2), 95–111. 10.1037/ccp0000761 [DOI] [PubMed] [Google Scholar]
  10. Bisby, M. A. , Scott, A. J. , Hathway, T. , Dudeney, J. , Fisher, A. , Gandy, M. , Heriseanu, A. I. , Karin, E. , Titov, N. , & Dear, B. F. (2022). Sudden gains in therapist‐guided versus self‐guided online treatments for anxiety or depression. Journal of Consulting and Clinical Psychology, 90(11), 861–871. 10.1037/ccp0000771 [DOI] [PubMed] [Google Scholar]
  11. Cai, H. , Xie, X.‐M. , Zhang, Q. , Cui, X. , Lin, J.‐X. , Sim, K. , Ungvari, G. S. , Zhang, L. , & Xiang, Y.‐T. (2021). Prevalence of suicidality in major depressive disorder: A systematic review and meta‐analysis of comparative studies. Frontiers in Psychiatry, 12, 690130. 10.3389/fpsyt.2021.690130 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chandra, S. S. , Bisby, M. A. , Scott, A. J. , Titov, N. , & Dear, B. F. (2025). The endometriosis pain course: A randomized controlled trial of an internet‐delivered psychological pain management program for endometriosis. Pain, 167, 80–93. 10.1097/j.pain.0000000000003784 [DOI] [PubMed] [Google Scholar]
  13. Cooper, M. , Messow, C.‐M. , McConnachie, A. , Freire, E. , Elliott, R. , Heard, D. , Williams, C. , & Morrison, J. (2018). Patient preference as a predictor of outcomes in a pilot trial of person‐centred counselling versus low‐intensity cognitive behavioural therapy for persistent sub‐threshold and mild depression. Counselling Psychology Quarterly, 31(4), 460–476. 10.1080/09515070.2017.1329708 [DOI] [Google Scholar]
  14. Cross, S. P. , & Alvarez‐Jimenez, M. (2024). The digital cumulative complexity model: A framework for improving engagement in digital mental health interventions. Frontiers in Psychiatry, 15, 1382726. 10.3389/fpsyt.2024.1382726 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Cuijpers, P. , Miguel, C. , Harrer, M. , Plessen, C. Y. , Ciharova, M. , Ebert, D. , & Karyotaki, E. (2023). Cognitive behavior therapy vs. control conditions, other psychotherapies, pharmacotherapies and combined treatment for depression: A comprehensive meta‐analysis including 409 trials with 52,702 patients. World Psychiatry, 22(1), 105–115. 10.1002/wps.21069 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cuijpers, P. , Reijnders, M. , & Huibers, M. J. H. (2019). The role of common factors in psychotherapy outcomes. Annual Review of Clinical Psychology, 15, 207–231. 10.1146/annurev-clinpsy-050718-095424 [DOI] [PubMed] [Google Scholar]
  17. Deady, M. , Collins, D. A. J. , Johnston, D. A. , Glozier, N. , Calvo, R. A. , Christensen, H. , & Harvey, S. B. (2022). The impact of depression, anxiety and comorbidity on occupational outcomes. Occupational Medicine, 72(1), 17–24. 10.1093/occmed/kqab142 [DOI] [PubMed] [Google Scholar]
  18. Dear, B. F. (2025). Credibility and expectations: Important factors for understanding clinical response, treatment completion, and dropout in internet‐delivered psychological interventions. Journal of Consulting and Clinical Psychology, 93(9), 595–608. 10.1037/ccp0000969 [DOI] [PubMed] [Google Scholar]
  19. Dear, B. F. , Titov, N. , Schwencke, G. , Andrews, G. , Johnston, L. , Craske, M. G. , & Mcevoy, P. (2011). An open trial of a brief transdiagnostic internet treatment for anxiety and depression. Behaviour Research and Therapy, 49(12), 830–837. 10.1016/j.brat.2011.09.007 [DOI] [PubMed] [Google Scholar]
  20. Devilly, G. J. , & Borkovec, T. D. (2000). Psychometric properties of the credibility/expectancy questionnaire. Journal of Behavior Therapy and Experimental Psychiatry, 31(2), 73–86. 10.1016/S0005-7916(00)00012-4 [DOI] [PubMed] [Google Scholar]
  21. GBD 2021 Diseases and Injuries Collaborators . (2024). Global incidence, prevalence, years lived with disability (YLDs), disability‐adjusted life‐years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990‐2021: A systematic analysis for the global burden of disease study 2021. Lancet (London, England), 403(10440), 2133–2161. 10.1016/S0140-6736(24)00757-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hansen, N. , Lambert, M. , & Forman, E. (2002). The psychotherapy dose‐response effect and its implications for treatment delivery services. Clinical Psychology: Science and Practice, 9, 329–343. [Google Scholar]
  23. Hedman‐Lagerlöf, E. , Carlbring, P. , Svärdman, F. , Riper, H. , Cuijpers, P. , & Andersson, G. (2023). Therapist‐supported internet‐based cognitive behaviour therapy yields similar effects as face‐to‐face therapy for psychiatric and somatic disorders: An updated systematic review and meta‐analysis. World Psychiatry, 22(2), 305–314. 10.1002/wps.21088 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Howard, K. I. , Kopta, S. M. , Krause, M. S. , & Orlinsky, D. E. (1986). The dose‐effect relationship in psychotherapy. American Psychologist, 41(2), 159–164. 10.1037/0003-066X.41.2.159 [DOI] [PubMed] [Google Scholar]
  25. Johnson, S. U. , Ulvenes, P. G. , Øktedalen, T. , & Hoffart, A. (2019). Psychometric properties of the GAD‐7 in a heterogeneous psychiatric sample. Frontiers in Psychology, 10, 1–8. 10.3389/fpsyg.2019.01713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Joubert, A. E. , Grierson, A. B. , Li, I. , Sharrock, M. J. , Moulds, M. L. , Werner‐Seidler, A. , Stech, E. P. , Mahoney, A. E. J. , & Newby, J. M. (2023). Managing rumination and worry: A randomised controlled trial of an internet intervention targeting repetitive negative thinking delivered with and without clinician guidance. Behaviour Research and Therapy, 168, 104378. 10.1016/j.brat.2023.104378 [DOI] [PubMed] [Google Scholar]
  27. Kambeitz‐Ilankovic, L. , Rzayeva, U. , Völkel, L. , Wenzel, J. , Weiske, J. , Jessen, F. , Reininghaus, U. , Uhlhaas, P. J. , Alvarez‐Jimenez, M. , & Kambeitz, J. (2022). A systematic review of digital and face‐to‐face cognitive behavioral therapy for depression. npj Digital Medicine, 5, 144. 10.1038/s41746-022-00677-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Karin, E. , Crane, M. F. , Dear, B. F. , Nielssen, O. , Heller, G. Z. , Kayrouz, R. , & Titov, N. (2021). Predictors, outcomes, and statistical solutions of missing cases in web‐based psychotherapy: Methodological replication and elaboration study. JMIR Mental Health, 8(2), e22700. 10.2196/22700 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kroenke, K. , Spitzer, R. L. , & Williams, J. B. W. (2001). The PHQ‐9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kumpasoğlu, G. B. , Campbell, C. , Saunders, R. , & Fonagy, P. (2025). Therapist and treatment credibility in treatment outcomes: A systematic review and meta‐analysis of clients' perceptions in individual and face‐to‐face psychotherapies. Psychotherapy Research, 35(1), 139–154. 10.1080/10503307.2023.2298000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Levis, B. , Benedetti, A. , & Thombs, B. D. (2019). Accuracy of patient health Questionnaire‐9 (PHQ‐9) for screening to detect major depression: Individual participant data meta‐analysis. BMJ, 365, l1476. 10.1136/bmj.l1476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Linardon, J. , Shatte, A. , Messer, M. , McClure, Z. , & Fuller‐Tyszkiewicz, M. (2023). Effects of Participant's choice of different digital interventions on outcomes for binge‐Spectrum eating disorders: A pilot doubly randomized preference trial. Behavior Therapy, 54(2), 303–314. 10.1016/j.beth.2022.09.007 [DOI] [PubMed] [Google Scholar]
  33. Luong, H. K. , Drummond, S. P. A. , & Norton, P. J. (2020). Elements of the therapeutic relationship in CBT for anxiety disorders: A systematic review. Journal of Anxiety Disorders, 76, 102322. 10.1016/j.janxdis.2020.102322 [DOI] [PubMed] [Google Scholar]
  34. Newman, M. G. , & Fisher, A. J. (2010). Expectancy/credibility change as a mediator of cognitive behavioral therapy for generalized anxiety disorder: Mechanism of action or proxy for symptom change? International Journal of Cognitive Therapy, 3, 245–261. 10.1521/ijct.2010.3.3.245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Owen, J. J. , Adelson, J. , Budge, S. , Kopta, S. M. , & Reese, R. J. (2016). Good‐enough level and dose‐effect models: Variation among outcomes and therapists. Psychotherapy Research, 26(1), 22–30. 10.1080/10503307.2014.966346 [DOI] [PubMed] [Google Scholar]
  36. Plummer, F. , Manea, L. , Trepel, D. , & McMillan, D. (2016). Screening for anxiety disorders with the GAD‐7 and GAD‐2: A systematic review and diagnostic metaanalysis. General Hospital Psychiatry, 39, 24–31. 10.1016/j.genhosppsych.2015.11.005 [DOI] [PubMed] [Google Scholar]
  37. Pontén, M. , Jonsjö, M. , Vadenmark, V. , Moberg, E. , Grannas, D. , Andersson, G. , Boersma, K. , Hedman‐Lagerlöf, E. , Kleinstaeuber, M. , Weise, C. , Kaldo, V. , Ljótsson, B. , Andersson, E. , Axelsson, E. , & Jensen, K. (2024). Association between expectations and clinical outcomes in online v. face‐to‐face therapy – An individual participant data meta‐analysis. Psychological Medicine, 54(6), 1207–1214. 10.1017/S0033291723003033 [DOI] [PubMed] [Google Scholar]
  38. Reins, J. A. , Boß, L. , Lehr, D. , Berking, M. , & Ebert, D. D. (2019). The more I got, the less I need? Efficacy of internet‐based guided self‐help compared to online psychoeducation for major depressive disorder. Journal of Affective Disorders, 246, 695–705. 10.1016/j.jad.2018.12.065 [DOI] [PubMed] [Google Scholar]
  39. Robinson, L. , Delgadillo, J. , & Kellett, S. (2020). The dose‐response effect in routinely delivered psychological therapies: A systematic review. Psychotherapy Research, 30(1), 79–96. 10.1080/10503307.2019.1566676 [DOI] [PubMed] [Google Scholar]
  40. Savir, T. , & Baumel, A. (2025). Capacity to invest effort as a predictor of preference for digital mental health interventions over psychotherapy: Cross‐sectional study using an ecological digital screening tool. Journal of Medical Internet Research, 27(1), e77802. 10.2196/77802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Schaffler, Y. , Probst, T. , Jesser, A. , Humer, E. , Pieh, C. , Stippl, P. , Haid, B. , & Schigl, B. (2022). Perceived barriers and facilitators to psychotherapy utilisation and how they relate to Patient's psychotherapeutic goals. Healthcare (Basel), 10(11), 2228. 10.3390/healthcare10112228 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Schleider, J. L. , Zapata, J. P. , Rapoport, A. , Wescott, A. , Ghosh, A. , Kaveladze, B. , Szkody, E. , & Ahuvia, I. L. (2025). Single‐session interventions for mental health problems and service engagement: Umbrella review of systematic reviews and meta‐analyses. Annual Review of Clinical Psychology, 21, 279–303. 10.1146/annurev-clinpsy-081423-025033 [DOI] [PubMed] [Google Scholar]
  43. Scholten, W. , Seldenrijk, A. , Hoogendoorn, A. , Bosman, R. , Muntingh, A. , Karyotaki, E. , Andersson, G. , Berger, T. , Carlbring, P. , Furmark, T. , Bouchard, S. , Goldin, P. , Kampmann, I. , Morina, N. , Kocovski, N. , Leibing, E. , Leichsenring, F. , Stolz, T. , van Balkom, A. , & Batelaan, N. (2023). Baseline severity as a moderator of the waiting list‐controlled Association of Cognitive Behavioral Therapy with Symptom Change in social anxiety disorder: A systematic review and individual patient data meta‐analysis. JAMA Psychiatry, 80(8), 822–831. 10.1001/jamapsychiatry.2023.1291 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sextl‐Plötz, T. , Steinhoff, M. , Baumeister, H. , Cuijpers, P. , Ebert, D. D. , & Zarski, A.‐C. (2024). A systematic review of predictors and moderators of treatment outcomes in internet‐ and mobile‐based interventions for depression. Internet Interventions, 37, 100760. 10.1016/j.invent.2024.100760 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Silva, S. , Barbosa, E. , Salgado, J. , & Cunha, C. (2021). Portuguese validation of the credibility/expectancy questionnaire in routine practice. Research in Psychotherapy: Psychopathology, Process, and Outcome, 24(1), 495. 10.4081/ripppo.2021.495 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Smeets, R. J. E. M. , Beelen, S. , Goossens, M. E. J. B. , Schouten, E. G. W. , Knottnerus, J. A. , & Vlaeyen, J. W. S. (2008). Treatment expectancy and credibility are associated with the outcome of both physical and cognitive‐behavioral treatment in chronic low back pain. The Clinical Journal of Pain, 24(4), 305–315. 10.1097/AJP.0b013e318164aa75 [DOI] [PubMed] [Google Scholar]
  47. Smith, S. M. , Dworkin, R. H. , Turk, D. C. , McDermott, M. , Eccleston, C. , Farrar, J. T. , Rowbotham, M. C. , Bhangwagar, Z. , Burke, L. B. , Cowan, P. , Ellenberg, S. S. , Evans, S. R. , Freeman, R. L. , Garrison, L. P. , Iyengar, S. , Jadad, A. R. , Jensen, M. P. , Junor, R. , Kamp, C. , … Wilson, H. D. (2020). Interpretation of chronic pain clinical trial outcomes: IMMPACT recommended considerations. Pain, 161(11), 2446–2461. 10.1097/j.pain.0000000000001952 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sperry, L. , & Binensztok, V. (2019). Ultra‐Brief Cognitive Behavioral Interventions: A New Practice Model for Mental Health and Integrated Care. Routledge. [Google Scholar]
  49. Spitzer, R. L. , Kroenke, K. , Williams, J. B. W. , & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder: The GAD‐7. Archives of Internal Medicine, 166(10), 1092–1097. 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  50. Staples, L. G. , Asrianti, L. , Karin, E. , Kayrouz, R. , Cross, S. , Bisby, M. , Fisher, A. , Dear, B. F. , Titov, N. , & Nielssen, O. (2022). Antidepressant medication use by patients accessing a national digital mental health service. Journal of Affective Disorders, 308, 305–313. 10.1016/j.jad.2022.04.042 [DOI] [PubMed] [Google Scholar]
  51. Tapp, C. , Scheurer, R. , Burgess, P. , Le, L. K.‐D. , Mihalopoulos, C. , Currier, D. , Pirkis, J. , & Harris, M. (2026). Uptake, utilisation and costs of treatment through better access from 2018 to 2022: An analysis of Medicare benefits schedule data. The Australian and New Zealand Journal of Psychiatry, 60(1), 11–24. 10.1177/00048674251409006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Thompson‐Hollands, J. , Bentley, K. H. , Gallagher, M. W. , Boswell, J. F. , & Barlow, D. H. (2014). Credibility and outcome expectancy in the unified protocol: Relationship to outcomes. Journal of Experimental Psychopathology, 5(1), 72–82. 10.5127/jep.033712 [DOI] [Google Scholar]
  53. Titov, N. , Dear, B. F. , McMillan, D. , Anderson, T. , Zou, J. , & Sunderland, M. (2011). Psychometric comparison of the PHQ‐9 and BDI‐II for measuring response during treatment of depression. Cognitive Behaviour Therapy, 40(2), 126–136. 10.1080/16506073.2010.550059 [DOI] [PubMed] [Google Scholar]
  54. Titov, N. , Dear, B. F. , Staples, L. G. , Terides, M. D. , Karin, E. , Sheehan, J. , Johnston, L. , Gandy, M. , Fogliati, V. J. , Wootton, B. M. , & McEvoy, P. M. (2015). Disorder‐specific versus transdiagnostic and clinician‐guided versus self‐guided treatment for major depressive disorder and comorbid anxiety disorders: A randomized controlled trial. Journal of Anxiety Disorders, 35, 88–102. 10.1016/j.janxdis.2015.08.002 [DOI] [PubMed] [Google Scholar]
  55. Upton, E. , Venkatesha, V. , Joubert, A. E. , Mahoney, A. E. J. , Moulds, M. L. , Werner‐Seidler, A. , & Newby, J. M. (2025). Mediators and predictors of treatment response in a brief online intervention for rumination and worry. Cognitive Therapy and Research, 49(5), 961–979. 10.1007/s10608-025-10593-2 [DOI] [Google Scholar]
  56. van Zwet, E. W. , Harrell, F. E., Jr. , & Senn, S. J. (2026). An empirical assessment of the cost of dichotomization of the outcome of clinical trials. Statistics in Medicine, 45, e70402. 10.1002/sim.70402 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wampold, B. E. , & Imel, Z. E. (2015). The great psychotherapy Debate (Vol. 53). https://www.taylorfrancis.com/books/mono/10.4324/9780203582015/great‐psychotherapy‐debate‐bruce‐wampold‐zac‐imel

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1. Univariate binary logistic regression output with complete case data.

Table S2. Multivariate binary logistic regression output for anxiety symptom models with complete case data.

Table S3. Multivariate binary logistic regression output for depression symptom models with complete case data.

BJC-65-1276-s001.docx (23.1KB, docx)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author pending ethical approval and institutional agreement. The data are not publicly available due to privacy or ethical restrictions.

Articles from The British Journal of Clinical Psychology are provided here courtesy of Wiley

RESOURCES